Jang et al.

Microbial Cell Factories 2011, 10:59

http://www.microbialcellfactories.com/content/10/1/59

RESEARCH Open Access

Retinoid production using metabolically

engineered Escherichia coli with a two-phase
culture system
Hui-Jeong Jang1†, Sang-Hwal Yoon1†, Hee-Kyung Ryu2, Jung-Hun Kim1, Chong-Long Wang1, Jae-Yean Kim1,
Deok-Kun Oh3 and Seon-Won Kim1*

Abstract
Background: Retinoids are lipophilic isoprenoids composed of a cyclic group and a linear chain with a hydrophilic
end group. These compounds include retinol, retinal, retinoic acid, retinyl esters, and various derivatives of these
structures. Retinoids are used as cosmetic agents and effective pharmaceuticals for skin diseases. Retinal, an
immediate precursor of retinoids, is derived by b-carotene 15,15’-mono(di)oxygenase (BCM(D)O) from b-carotene,
which is synthesized from the isoprenoid building blocks isopentenyl diphosphate (IPP) and dimethylallyl
diphosphate (DMAPP). Retinoids are chemically unstable and biologically degraded via retinoic acid. Although
extensive studies have been performed on the microbial production of carotenoids, retinoid production using
microbial metabolic engineering has not been reported. Here, we report retinoid production using engineered
Escherichia coli that express exogenous BCM(D)O and the mevalonate (MVA) pathway for the building blocks
synthesis in combination with a two-phase culture system using a dodecane overlay.
Results: Among the BCM(D)O tested in E. coli, the synthetic retinoid synthesis protein (SR), based on
bacteriorhodopsin-related protein-like homolog (Blh) of the uncultured marine bacteria 66A03, showed the highest
b-carotene cleavage activity with no residual intracellular b-carotene. By introducing the exogenous MVA pathway,
8.7 mg/L of retinal was produced, which is 4-fold higher production than that of augmenting the MEP pathway
(dxs overexpression). There was a large gap between retinal production and b-carotene consumption using the
exogenous MVA pathway; therefore, the retinal derivatives were analyzed. The derivatives, except for retinoic acid,
that formed were identified, and the levels of retinal, retinol, and retinyl acetate were measured. Amounts as high
as 95 mg/L retinoids were obtained from engineered E. coli DH5a harboring the synthetic SR gene and the
exogenous MVA pathway in addition to dxs overexpression, which were cultured at 29°C for 72 hours with 2YT
medium containing 2.0% (w/v) glycerol as the main carbon source. However, a significant level of intracellular
degradation of the retinoids was also observed in the culture. To prevent degradation of the intracellular retinoids
through in situ extraction from the cells, a two-phase culture system with dodecane was used. The highest level of
retinoid production (136 mg/L) was obtained after 72 hours with 5 mL of dodecane overlaid on a 5 mL culture.
Conclusions: In this study, we successfully produced 136 mg/L retinoids, which were composed of 67 mg/L
retinal, 54 mg/L retinol, and 15 mg/L retinyl acetate, using a two-phase culture system with dodecane, which
produced 68-fold more retinoids than the initial level of production (2.2 mg/L). Our results demonstrate the
potential use of E. coli as a promising microbial cell factory for retinoid production.

* Correspondence: swkim@gnu.ac.kr
† Contributed equally
1
Division of Applied Life Science (BK21 Program), PMBBRC, Gyeongsang
National University, Jinju 660-701, Korea
Full list of author information is available at the end of the article

© 2011 Jang et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons
Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
Jang et al. Microbial Cell Factories 2011, 10:59 Page 2 of 12
http://www.microbialcellfactories.com/content/10/1/59

Background retinoids are the principal commercial source. Retinol

Retinoids are a class of lipophilic isoprenoid molecules
that are related chemically to vitamin A [1]. Retinoids are
composed of a b-ionone ring and a polyunsaturated side
chain, with an alcohol (retinol), an aldehyde (retinal), a
carboxylic acid (retinoic acid), or an ester (retinyl esters)
functional group (Figure 1A). They play an essential role
in vision, bone development, reproduction, and skin
health as antioxidants and are also known to reduce the
risk of certain cancers. Retinoids have attracted increased
attention in recent years as cosmetic active ingredients
and effective pharmaceuticals for skin diseases [2]. The
retinoid market size has been estimated to be about 1.6
billion dollars worldwide. Chemically synthesized
has been produced from acidification or hydrolysis of ret-
inal that is synthesized chemically by the reduction of a
pentadiene derivative [1,3]. However, these chemical pro-
cesses have some disadvantages, such as complex purifi-
cation steps and the formation of undesirable by-
products. Animals produce retinoids from carotenoids
(the most effective being b-carotene) obtained from fruits
and vegetables, but plants cannot synthesize retinoids.
The retinoid synthesis pathway is present only in micro-
organisms containing bacteriorhodopsin or proteorho-
dopsin with retinal as a prosthetic group [4,5]. However,
the microorganisms produce the protein-bound form of
retinal and are not appropriate for mass production of

Figure 1 Retinoid biosynthesis. The conversion of b-carotene to retinoids, including retinal, retinol, retinoic acid, and retinyl esters (A).
Biosynthesis of retinal in Escherichia coli using the MEP pathway and the exogenous MVA pathway (B). The gene names and the encoded
enzymes are as follows: atoB/phaA, acetoacetyl-CoA synthase; mvaS, HMG-CoA synthase; mvaA, HMG-CoA reductase; mvaK1, mevalonate kinase;
mvaK2, phosphomevalonate kinase; mvaD, mevalonate-5-diphosphate decarboxylase; idi, IPP isomerase; dxs, DXP synthase; ispA, FPP synthase;
crtE, GGPP synthase; crtB, phytoene synthase; crtI, phytoene desaturase; crtY, lycopene cyclase; BCM(D)O, b-carotene 15,15’-mono- or dioxygenase.
Pathway intermediates are as follows: G3P, D-glyceraldehyde-3-phosphate; DXP, 1-deoxy-D-xyluose-5-phosphate; MEP, 2-C-methyl-D-erythritol-4-
phosphate; IPP, isopentenyl diphosphate; FPP, farnesyl diphosphate; GGPP, geranylgeranyl diphosphate.
Jang et al. Microbial Cell Factories 2011, 10:59
http://www.microbialcellfactories.com/content/10/1/59
free retinoids. To date, only a few attempts of the biologi-
cal production of retinoids have been reported [5,6]. It
would be beneficial to develop a biotechnological process
for retinoid production using a metabolically engineered
microorganism.
Retinoids are composed of 3 isopentenyl diphosphate
(IPP) units and 1 dimethylallyl diphosphate (DMAPP),
which are the common five-carbon building blocks of all
isoprenoids. The IPP and DMAPP building blocks are
generally synthesized via the 2-C-methyl-D-erythritol-4-
phosphate (MEP) and mevalonate (MVA) pathways in
prokaryotes and eukaryotes, respectively [7-9]. Recombi-
nant E. coli harboring an exogenous MVA pathway has
been used for the successful production of isoprenoids,
such as amorphadiene, carotenoids, and farnesol [10-14].
In particular, we reported the high-level production of b-
carotene (465 mg/L) from E. coli harboring an engi-
neered MVA pathway [13,15]. The recombinant E. coli
can be engineered to produce retinal by introducing b-
carotene 15,15’-mono(di)oxygenase (BCM(D)O) as a b-
carotene cleavage enzyme (Figure 1B).
The cleavage of b-carotene by BCM(D)O (E.C.
1.13.11.21 or E.C. 1.14.99.36) is the initial key step of
synthesis of various retinoids from b-carotene. The clea-
vage reactions can be classified as central and eccentric.
In the central cleavage, BCM(D)O cleaves the central
double bond (15, 15’) of the polyene chain of b-carotene
to yield two molecules of retinal. In the eccentric clea-
vage, BCM(D)O randomly cleaves any double bond in
the polyene chain to produce b-apo-carotenals with dif-
ferent side chain lengths. In the cleavage reactions,
BCMO utilizes an oxygen atom derived from molecular
oxygen and water via an epoxide intermediate, whereas
BCDO employs a molecular oxygen via an unstable
dioxetane intermediate [16,17]. Retinal is converted to
retinol and retinoic acid by retinol dehydrogenase and
retinal dehydrogenase/oxidase, respectively (Figure 1A)
[18,19]. Retinol is esterified to retinyl esters by retinol
acyltransferase [20].
Retinoids are chemically unstable and readily oxidized
and isomerized by heat, oxygen, and light due to their
reactive conjugated double bonds [21,22]. Biologically,
retinoids are also easily degraded via retinoic acid. The
oxidative degradation begins with the conversion of reti-
noic acid to more polar metabolites, such as 4-hydroxy-
and 4-oxo-retinoic acids [23,24]. Therefore, successful
production of retinoids can be achieved by preventing
both chemical and biological degradation. To our knowl-
edge, retinoid production using metabolically engineered
microorganisms has never been reported. In this study,
we cloned the BCM(D)O genes from several organisms
and introduced each gene into recombinant E. coli that
produce b-carotene. An exogenous MVA pathway was
also utilized to increase retinoid production. A two-phase
Page 3 of 12
culture system using a dodecane layer over the culture
broth was found to minimize the intracellular degrada-
tion of retinoids through in situ extraction from the cells.
Results
Comparison of retinal production from various BCM(D)O
genes
Retinal can be produced by introducing the b-carotene
mono(di)oxygenase (BCM(D)O) gene into recombinant
E. coli that produces b-carotene. We cloned the BCM
(D)O genes from two bacteria, Halobacterium sp. NRC-
1 (blh and brp) and Natronomonas pharaonis (brp2),
and a vertebrate, Mus musculus (Bcmo1). We also
synthesized a codon-optimized BCDO gene (SR) based
on the amino acid sequence of the uncultured marine
bacterium 66A03 blh gene. The BCM(D)O genes were
used to construct the retinal synthesis plasmids pT-
HBblh, pT-HBbrp, pT-HBbrp2, pT-HBBcmo1 and
pT-HBSR, respectively. The recombinant E. coli cells har-
boring each retinal plasmid were cultured in 2YT medium
containing 0.5% (w/v) glycerol and 0.2% (w/v) arabinose as
carbon sources for 48 hours at 29°C (Figure 2). The cells
were cultured without IPTG induction because leaky
expression in the 2YT complex medium was sufficient
for retinal production. We have previously observed
severe inhibition of both cell growth and carotenoid pro-
duction due to IPTG induction [12], and similar results
were obtained in this retinal production study (data not
shown). Recombinant E. coli harboring pT-HBblh, pT-
HBbrp, or pT-HBSR produced 2.2, 0.8, or 1.4 mg/L ret-
inal after 24 hours, respectively. However, retinal produc-
tion from E. coli containing pT-HBblh or pT-HBbrp
decreased to 0.7 or 0.4 mg/L after 48 hours, respectively,
whereas retinal production of E. coli (pT-HBSR)
increased slightly. The decreases in retinal production
after 24 hours may be due to intracellular oxidative
degradation of retinal. The amounts of retinal obtained
from the culture should be dependent on both intracellu-
lar synthesis and degradation. Retinal production rates in
E. coli harboring pT-HBblh or pT-HBbrp might be lower
than their degradation rates after 24 hours of the culture.
Trace amounts of retinal were detected in the culture of
E. coli harboring pT-HBbrp2 or pT-HBBcmo1. E. coli
(pT-HB) with no BCM(D)O gene produced 35 mg/L b-
carotene and no retinal. If there is no retinal degradation,
b-carotene consumption amounts by BCM(D)Os are
exactly proportional to retinal production because b-car-
otene is the immediate precursor of retinal. The b-caro-
tene cleavage activity of SR was suspected to be the
highest among the tested BCM(D)Os because b-carotene
remained in the culture of E. coli containing BCM(D)Os
other than SR. The SR enzyme was therefore selected for
retinal production in further experiments. Cell growth
was not affected by overexpression of the BCM(D)O
Jang et al. Microbial Cell Factories 2011, 10:59 Page 4 of 12
http://www.microbialcellfactories.com/content/10/1/59

dxs gene was introduced to augment the MEP pathway

in pT-HBSR, resulting in pT-DHBSR (Figure 3). Retinal
production from E. coli (pT-DHBSR) was a slightly
higher than that of E. coli (pT-HBSR) after 24 hours,
but there was no difference after 48 hours, whereas the
b-carotene production from E. coli (pT-DHB) was 1.5-
fold higher due to dxs overexpression compared with
E. coli (pT-HB). The exogenous MVA pathway in E. coli
has been shown to dramatically increase isoprenoid pro-
duction by providing sufficient amounts of the IPP and
DMAPP building blocks [10,13]. E. coli (pT-DHBSR/pS-
NA) harboring the exogenous MVA pathway produced
8.7 mg/L retinal after 48 hours, which was 4-fold higher
than that of E. coli (pT-DHBSR). In E. coli strains con-
taining the SR gene, little or no b-carotene remained in
the cells probably due to b-carotene cleavage by SR.
There was large gap between the amount of b-carotene
(substrate) consumed and the amount of retinal

Figure 2 Comparison of retinal production from various BCM

(D)O genes. Retinal and b-carotene production and cell growth of
E. coli harboring pT-HB, pT-HBblh, pT-HBbrp, pT-HBbrp2, pT-
HBBCMO1, and pT-HBSR. The blh gene of Halobacterium sp. NRC-1,
the brp gene of Halobacterium sp. NRC-1, the brp2 gene of N.
pharaonis, the BCMO1 gene of M. musculus and the SR codon-
optimized blh gene of the uncultured marine bacterium 66A03
were introduced into the b-carotene plasmid, pT-HB, resulting in pT-
HBblh, pT-HBbrp, pT-HBbrp2, pT-HBBCMO1, and pT-HBSR,
respectively. Culture was carried out in 2YT medium containing
0.5% (w/v) glycerol and 0.2% (w/v) arabinose for 48 hours at 29°C.
Open bars and solid bars represent 24 hrs and 48 hrs, respectively.

genes, except for the N. pharaonis brp gene, which

showed growth retardation.

Engineering the MEP and MVA pathways to supply

building blocks
The retinal building blocks of IPP and DMAPP can be
synthesized in E. coli via the endogenous MEP pathway
Figure 3 Retinal production using the MEP and MVA pathways.
and the exogenous MVA pathway (Figure 1B). The Retinal production, b-carotene production, and cell growth of E. coli
synthesis of 1-deoxy-d-xylulose-5-phosphate (DXP) has harboring pT-HB, pT-HBSR, pT-DHB, and pT-DHBSR and E. coli
been reported as the critical rate-limiting step in the harboring pT-DHB, or pT-DHBSR with the MVA pathway plasmid of
MEP pathway. Thus, overexpression of DXP synthase pS-NA. Culture was carried out in 2YT medium containing 0.5% (w/
v) glycerol and 0.2% (w/v) arabinose for 48 hours at 29°C. Open
(encoded by dxs) increased the levels of lycopene and b-
bars and solid bars represent 24 hrs and 48 hrs, respectively.
carotene production in our previous study [15,25]. The
Jang et al. Microbial Cell Factories 2011, 10:59
http://www.microbialcellfactories.com/content/10/1/59
(product) produced. We hypothesized that there are
other cellular reactions that metabolize retinal in E. coli
besides the biological degradation of retinal. The forma-
tion of other retinoids derived from retinal by promiscu-
ous enzymes in E. coli was considered. Because retinal
could be converted into retinol, retinoic acid, and retinyl
ester by cellular enzymatic reactions (Figure 1A), these
retinal derivatives were analyzed in the E. coli cultures.
Formation of the derivatives, except for retinoic acid,
was found, and the levels of retinal, retinol, and retinyl
acetate produced were measured in further analyses.
Effects of E. coli strains, culture conditions and carbon
sources on retinoid production
The effect of the E. coli strain used on the production of
retinoids, including retinal, retinol, and retinyl acetate,
was investigated. Isoprenoid production has been shown
to significantly depend on the E. coli strain used
[13,25,26]. Therefore, retinoid production was analyzed
in five E. coli strains, MG1655, DH5a, XL1-Blue, S17-1,
and BL21 (DE3), harboring pT-DHBSR and pS-NA
(Additional file 1). E. coli DH5a exhibited the highest
level of retinoid production (40 mg/L) after 36 hours,
followed by E. coli S17-1 and XL1-Blue, which produced
approximately 22 mg/L retinoids. However, small
amounts of retinoids were obtained from E. coli
MG1655 and BL21 (DE3). Thus, E. coli DH5a was
selected as the optimal strain for retinoid production.
The effect of dissolved oxygen on retinoid production
was investigated with different working volumes in a 30
mm diameter test tube (Additional file 2). Retinoid pro-
duction reached the maximum level earlier with a lower
working volume (which corresponds to higher dissolved
oxygen) and began to decline earlier, probably due to
oxidative degradation. In a 10 mL working volume, both
cell growth and retinoid production were retarded, but
less product degradation was observed. The optimal
working volume for retinoid production was found to
be 7 mL. Retinoid production was also affected by cul-
ture temperature (Additional file 3), and the highest
production was obtained at 29°C. These results agreed
with the optimal strains and culture conditions for b-
carotene production found in our previous study [15],
which was expected because b-carotene was the
immediate precursor of retinal. Retinoid production on
different carbon sources was compared (Additional file
4). Glycerol was the best carbon source for retinoid pro-
duction. If glucose or galactose was used as the carbon
source, the production of retinoids was lower than that
without a carbon source. Therefore, the effects of the
glycerol concentration on retinoid production and cell
growth were investigated. E. coli DH5a (pT-DHBSR/pS-
NA) was grown in 2YT medium containing 0.0% to
2.0% (w/v) glycerol at 29°C (Figure 4). Cell growth
Page 5 of 12
increased with increasing glycerol concentrations, and
stationary phase was reached at 36, 48, and 72 hours in
media containing glycerol concentrations of 0.5, 1.0, and
2.0% (w/v), respectively. Retinoid production was maxi-
mal at these times and then significantly decreased dur-
ing stationary phase. Retinoid production appeared to
increase mainly after 24 hours. The highest level of reti-
noid production (95 mg/L) was obtained in 2.0% (w/v)
glycerol among the concentrations tested, which was
2.4-fold higher than the maximal retinoid production
with 0.5% (w/v) glycerol. A low level of retinoids was
produced with no addition of glycerol. The increase in
glycerol concentration retarded stationary phase growth
and prolonged the period of retinoid production but
failed to prevent retinoid degradation. The retinoids
obtained in the culture consisted of retinal, retinol, and
retinyl acetate, and retinol was the major component of
the retinoids. Retinol and retinyl acetate are known to
be formed by retinol dehydrogenase and retinol acyl-
transferase, respectively. It was very interesting that reti-
nol and retinyl acetate were produced in E. coli without
introduction of the retinol dehydrogenase and retinol
acyltransferase genes. In all of the cultures, a severe
decrease in retinoid production was observed during
stationary phase growth, which might be due to a lack
of additional retinoid synthesis and intracellular oxida-
tive degradation during stationary phase.
Two-phase culture using dodecane for in situ extraction
of retinoids
To prevent intracellular retinoid degradation, a two-
phase culture system using the hydrophobic solvent
dodecane was performed for in situ extraction of reti-
noids from the cells. Dodecane was chosen for its low
toxicity to E. coli [27], high hydrophobicity (log PO/W,
6.6) for the extraction of hydrophobic retinoids, and low
volatility, which prevents loss due to evaporation. A
two-phase culture system using decane has been suc-
cessfully applied to lycopene production [28].
In this study, 1 mL of dodecane was layered over 5
mL of culture broth (Figure 5). Retinoids were extracted
into the dodecane phase, and negligible amounts of reti-
noids were detected in the cell mass and culture broth
(data not shown). As a result, retinoid production was
measured only from the dodecane phase. The in situ
extraction by dodecane could minimize intracellular
degradation of the retinoids. The retinoids in the dode-
cane phase seemed to be relatively stable and were not
subject to significant oxidative degradation. Compared
with the results in Figure 4 (no dodecane overlay), reti-
noid production in the two-phase system with 1 mL of
dodecane overlay was significantly higher even at 24
hours, and no decrease in retinoid production was
observed during stationary phase, while cell growth was
Jang et al. Microbial Cell Factories 2011, 10:59 Page 6 of 12
http://www.microbialcellfactories.com/content/10/1/59

Figure 4 Effect of glycerol concentration on retinoid production. Retinoid (retinal, retinol, and retinyl acetate) production and cell growth of
E. coli harboring pT-DHBSR and pS-NA. Culture was carried out in 2YT medium containing 0%, 0.5%, 1%, and 2% (w/v) glycerol and 0.2% (w/v)
arabinose at 29°C. For retinoid production, retinal, retinol, and retinyl acetate are indicated with light gray, dark gray, and black, respectively. For
cell growth, the glycerol concentrations are indicated as symbols; 0%, squares; 0.5%, circles; 1%, triangles; 2%, reversed triangles.

Figure 5 Effect of dodecane overlay on retinoid production. Retinoid production and cell growth of E. coli (pT-DHBSR/pS-NA) in two-phase
culture system using a 1 mL of dodecane overlay and 5 mL of culture broth. The culture was carried out in 2YT medium containing 0.5%, 1%,
and 2% (w/v) glycerol and 0.2% (w/v) arabinose at 29°C. For retinoid production, retinal, retinol, and retinyl acetate are indicated with light gray,
dark gray, and black, respectively. For cell growth, the glycerol concentrations are indicated as symbols; 0.5%, squares; 1%, circles; 2%, triangles.
Jang et al. Microbial Cell Factories 2011, 10:59
http://www.microbialcellfactories.com/content/10/1/59
not affected by the dodecane overlay. However, retinoid
production in the culture containing 2% (w/v) glycerol
was not higher than that obtained with 1% (w/v) gly-
cerol, although cell growth was significantly enhanced
with the higher glycerol concentration. The dodecane
overlay volume of 1 mL might be insufficient for effec-
tive in situ extraction of retinoids in cultures containing
2% (w/v) glycerol.
To investigate the effect of the dodecane overlay
volume on retinoid production and cell growth, 1 mL to
5 mL of dodecane were initially overlaid on cultures con-
taining 2% (w/v) glycerol (Figure 6A). Total retinoid pro-
duction was enhanced as the dodecane overlay volume
increased. The highest retinoid production of 136 mg/L
was obtained after 72 hours of culture with 5 mL dode-
cane, which was about 2-fold higher than that with 1 mL
of dodecane (65 mg/L). Culture times longer than 72
hours with 5 mL of dodecane showed no further increase
in retinoid production, which also remained at the maxi-
mum level without degradation (data not shown). The
dodecane overlay volume was increased to 6 mL in the
culture by addition of 2 mL of dodecane at 0, 24, and 48
hours. In the 6 mL dodecane overlaid culture, there was
no increase in the total retinoid production compared
with the culture with 5 mL of dodecane. Retinoid pro-
duction also did not increase in a culture with an initial
overlay of 6 mL of dodecane (data not shown). Cell
growth in all of the cultures with dodecane was slightly
higher than that without dodecane (Figure 6A).
The proportions of the retinoids obtained with the var-
ious dodecane overlay volumes were determined (Figure
6B). An outstanding difference in the proportions of ret-
inal and retinol obtained with and without the dodecane
overlays was found. The proportion of retinal among the
retinoids after 48 hours was approximately 51% (w/w) in
the dodecane overlaid cultures and 23% in the culture
without dodecane overlay, whereas the retinol proportion
was 30% to 39% in the dodecane overlaid cultures and
59% in the culture without dodecane overlay. Therefore,
the dodecane overlay increased the proportion of retinal
but decreased the proportion of retinol. Retinal seemed
to be extracted from cells by dodecane before it could be
intracellularly converted into retinol, as retinol is formed
from retinal in cells. The proportion of retinyl acetate
after 48 hours was below 20% in both the dodecane over-
laid culture and culture without overlay, which was lower
than those of retinal and retinol. In the dodecane-over-
laid cultures, the proportion of retinyl acetate decreased
with increasing culture time, suggesting that retinyl acet-
ate formation decreased during culture. We concluded
that the dodecane overlay prevented the decrease in reti-
noid production during stationary phase growth and
increased retinoid production.
Page 7 of 12
Discussion
E. coli harboring the synthetic BCDO (SR) gene, which
was the codon-optimized blh gene from the uncultured
marine bacterium 66A03, successfully produced retinal
from b-carotene. Interestingly, the E. coli also produced
retinol and retinyl acetate. We hypothesize that promis-
cuous enzymes in E. coli are able to metabolize retinal
to produce retinol and retinyl acetate. Retinal is metabo-
lized to retinol and retinyl acetate in a sequential man-
ner by retinol dehydrogenase and retinol acyltransferase,
respectively. Therefore, we investigated a presence of a
potential retinol dehydrogenase in E. coli. The ybbO
gene in E. coli 83972 (Accession No. ZP_04002297) was
identified as a possible retinol dehydrogenase in the
NCBI protein database, although we did not expect to
find specific enzymes that metabolize foreign com-
pounds, such as retinoids. The ybbO gene is annotated
in E. coli strain MG1655 as a predicted oxidoreductase
and has the highest identity and similarity (31% and
52%, respectively) in the E. coli genome to H. sapiens
retinol dehydrogenase (Accession No. AAC72923) based
on the BLASTP analysis of NCBI http://www.ncbi.nlm.
nih.gov/blast/. It has been reported through sequence
comparisons and phylogenetic analysis that the ybbO
gene may be an example of horizontal gene transfer
from a eukaryotic retinol dehydrogenase ancestor [29].
To identify putative homologues of retinol acyltransfer-
ase in E. coli, BLASTP analysis was performed with reti-
nol acyltransferases of H. sapiens (Accession No.
NP_004735) and two bacteria, Shewanella putrefaciens
CN-32 (YP_001185024) and Trichodesmium erythraeum
IMS 101 (YP_723688), because the protein sequence of
bacterial retinol acyltransferase was available in only
these two bacteria. No homologue of retinol acyltrans-
ferase was identified from the BLASTP analysis. An
alternative approach for the identification of a homolo-
gous gene would be to delete the genes of all acyltrans-
ferases present in E. coli. Biological degradation of
retinoids is initiated from retinoic acid. We observed
significant intracellular degradation of retinoids during
stationary phase growth. If retinoic acid is quickly
degraded in E. coli, it would not be detected in the cul-
ture during retinoid production. We hypothesize that
retinoic acid is formed in our E. coli strain engineered
to produce retinoids. Retinal is converted to retinoic
acid by retinal dehydrogenase. Salmonella enterica is
known to have a retinal dehydrogenase (Accession No.
CBY96723). BLASTP analysis was performed on the E.
coli genome with the retinal dehydrogenase of S. enter-
ica, and eutE (predicted aldehyde dehydrogenase/etha-
nolamine utilization protein) was identified as a
homologue (with 94% identity and 97% similarity). The
retinal dehydrogenase of H. sapiens (Accession No.
Jang et al. Microbial Cell Factories 2011, 10:59 Page 8 of 12
http://www.microbialcellfactories.com/content/10/1/59

Figure 6 Effect of dodecane overlay volume on retinoid production. Effect of the dodecane overlay volume on retinoid production and cell
growth of E. coli (pT-DHBSR/pS-NA) in the two-phase culture. The culture was carried out in 2YT medium containing 2% (w/v) glycerol and 0.2%
(w/v) arabinose at 29°C, and different dodecane volumes from 1 mL to 5 mL were overlaid on the 5 mL culture broth. For the 6 mL dodecane
overlay, the dodecane overlay was divided into three aliquots of 2 mL added at 0, 24 and 48 hours. For retinoid production, retinal, retinol, and
retinyl acetate are indicated with light gray, dark gray, and black, respectively. For cell growth, the overlaid dodecane volumes are indicated as
symbols; 0 mL, closed squares; 1 mL, closed circles; 2 mL, closed triangles; 3 mL, closed reversed triangles; 4 mL, closed diamonds; 5 mL, open
squares; 6 mL, open circles (A). Proportions of retinoids produced as a function of culture time and dodecane overlay volume are represented as
percentages of the total retinoids. Retinal, retinol, and retinyl acetate are indicated in light gray, dark gray, and black, respectively (B).
Jang et al. Microbial Cell Factories 2011, 10:59
http://www.microbialcellfactories.com/content/10/1/59
NP_733798) was also used for the same BLASTP analy-
sis, and puuC (gamma-Glu-gamma-aminobutyraldehyde
dehydrogenase) was found to have the highest homology
(42% identity and 63% similarity) to the retinal dehydro-
genase. If eutE or puuC are involved in the formation of
retinoic acid, deletion of these genes will prevent the
biological degradation of retinoids via retinoic acid,
resulting in an enhancement in total retinoid
production.
In the cultures without a dodecane overlay, there was
a significant decrease in retinoid production during sta-
tionary phase growth. This might be due to increased
oxidative degradation of retinoids by reactive oxygen
species, such as hydrogen peroxide and superoxide,
which are generated at high levels during stationary
phase. Retinoids are easily oxidized as antioxidants by
reactive oxygen species. Oxidative retinoid degradation
could be decreased by overexpression of catalases (Kat
E/G) and superoxide dismutases (Sod A/B/C), which
scavenge reactive oxygen species. To prevent biological
degradation of retinoids inside of the cells, in situ
extraction of retinoids was performed with a two-phase
culture system using dodecane. Oxidizable compounds
are easily oxidized and degraded by molecular oxygen
dissolved in the aqueous phase, whereas compounds in
hydrophobic solvents, including dodecane, are seques-
tered and more stable [30,31]. Thus, dodecane was
found to efficiently extract hydrophobic retinoids from
cells and preserve the products during two-phase cul-
ture. In a previous report, a two-phase culture system
using decane was successfully applied for lycopene pro-
duction [28]; however, lycopene was inefficiently
extracted from recombinant E. coli without partial diges-
tion of the cell wall by lysozyme. In this study, the use
of lysozyme for cell wall digestion was not required for
the in situ extraction of retinoids. Retinoids are effi-
ciently released from cells without removing the cell
walls because retinoids (C20, isoprenoid molecule) are
half the size of lycopene (C40). In the two-phase culture
for retinoid production, b-carotene should be retained
inside of the cells because it is the immediate precursor
of retinoids. If it is extracted in the dodecane phase, it
would not be available for the cleavage reaction by BCM
(D)O located in the cytosol. Even though extraction of
b-carotene by dodecane would not be expected because
it is a C40 carotenoid like lycopene, two-phase culture
for b-carotene production was performed to confirm
that b-carotene is retained in the cells (Additional file
5). A negligible amount of b-carotene was detected in
the dodecane phase and almost all of the b-carotene
was retained in the cells. There was no significant differ-
ence in both b-carotene production and cell growth
between cultures with and without a dodecane overlay.
The two-phase culture system prevents intracellular
Page 9 of 12
degradation of retinoids and provides a driving force for
further retinoid production. Physical sequestration of
the product from a reaction system drives the reaction
to high efficiency without the effects of product inhibi-
tion or reaction equilibrium [32,33]. In the two-phase
culture system, the retinoids were sequestered from the
cells in the dodecane phase, and production was
enhanced. A total retinoid production of 122 mg/L was
obtained after 48 hours in a culture with a 5 mL dode-
cane overlay, whereas half of this amount (60 mg/L) was
produced after 48 hours without a dodecane overlay.
Thus, the dodecane-overlaid two-phase culture system
could be employed for other engineered systems that
produce lipophilic small molecules.
Conclusions
Our results represent the first report on retinoid bio-
synthesis using metabolically engineered E. coli. In this
study, we successfully produced 136 mg/L retinoids,
which were composed of retinal (67 mg/L), retinol (54
mg/L), and retinyl acetate (15 mg/L), using a two-phase
culture system with dodecane, which was a 68-fold
improvement from the initial level of retinoid produc-
tion (2.2 mg/L). This improvement was achieved with
use of (1) an efficient marine bacterial BCDO gene that
was codon-optimized for expression in E. coli, (2) intro-
duction of an exogenous MVA pathway that successfully
provided the building blocks IPP and DMAPP for reti-
noid synthesis, and (3) the use of a two-phase culture
system with dodecane that prevented intracellular degra-
dation of the retinoids and provided a driving force for
retinoid production. Retinal, retinol and retinyl acetate
were contained in the retinoids produced from the
recombinant E. coli, which suggests that E. coli has the
potential to synthesize multiple retinoids, which can be
used for different commercial applications. Based on
this potential, the retinoid synthesis pathway of E. coli
can be reengineered to produce a specific retinoid
through elaborative genetic manipulations, such as gene
deletions and overexpression of genes involved in the
modification of retinoids. Therefore, E. coli is a geneti-
cally tractable host that is a promising microbial cell
factory for the engineered production of retinoids.
Methods
Bacterial strains and culture conditions
The bacterial strains used in this study are listed in
Table 1. E. coli DH5a was used for gene cloning and
retinoid production. E. coli strains MG1655, BL21
(DE3), XL1-Blue and S17-1 were candidate host strains
for retinoid production (Table 1). Culture for retinoid
production was carried out in 2YT medium (16 g tryp-
tone, 10 g yeast extract, and 5 g NaCl per liter) using a
shaking incubator at 29°C and 250 rpm. Glycerol and
Jang et al. Microbial Cell Factories 2011, 10:59 Page 10 of 12
http://www.microbialcellfactories.com/content/10/1/59

Table 1 Strains, plasmids and primers used in this study

Strains, plasmids and Description Reference or
primers source
E. coli strains
MG1655 K12, Wild type
DH5a F-, j80dlacZΔM15, Δ(lacZYA-argF)U169, deoR, recA1 endA1, hsdR17(rK- mK+), phoA, supE44, l-, thi-1,
gyrA96, relA1
XL1-Blue hsdR17, supE44, recA1, endA1, gyrA46, thi, relA1, lac/F’[proAB+, lacIq, lacZΔM15::Tn10(tetr)]
S17-1 recA pro hsdR RP4-2-Tc::Mu-Km::Tn7
BL21(DE3) F-, ompT, hsdSB(rB-mB-), gal (lc I857, ind1, Sam7, nin5, lacUV5-T7gene1), dcm (DE3)

Plasmids
pBluescript Plac cloning vector, ColE1 origin, lacZ, Ampr Stratagene
pSTV28 Plac expression vector, pACYC184 origin, lacZ, Cmr Takara
pTrc99A Ptrc expression vector, pBR322 origin, lacIq, Cmr Amersham
Bioscience
pT-HB pTrc99A containing crtE, crtB, and crtI from P. agglomerans, crtY from P. ananatis, and ipiHP1 from H. [15]
pluvialis
pT-DHB pT-HB containing dxs from E. coli [15]
pT-HBblh pT-HB containing blh from Halobacterium sp. NRC-1 This study
pT-HBbrp pT-HB containing brp from Halobacterium sp. NRC-1 This study
pT-HBbrp2 pT-HB containing brp2 from N. pharaonis This study
pT-HBBcmo1 pT-HB containing Bcmo1 from M. musculus This study
pT-HBSR pT-HB containing the codon-optimized blh gene (SR) from uncultured marine bacterium 66A03 This study
pT-DHBSR pT-DHB containing the codon-optimized blh gene (SR) from uncultured marine bacterium 66A03 This study
pS-NA pSTV28 containing mvaE and mvaS from E. faecalis; mvaK1, mvaK2, and mvaD from S. pneumoniae; [13]
and idi from E. coli

Primersa
blhE-F 5’-GGAATTCAGGAGGTGTTCGGCATGCCACACGG-3’
blh-R 5’-GACTAGTTAGAGGACGCCCTGCACGCGGTC-3’
brpE-F 5’-GGAATTCAGGAGGTATTCATATGAGCAATAGGTC-3’
brp-R 5’-GACTAGTTATGGGACGTACCAGATGCCG-3’
brp2E-F 5’-GGAATTCAGGAGGCCGAGTATGAGTAACGCGTC-3’
brp2-R 5’-GACTAGTTATGCTCCGGGTCGCCAGAG-3’
Bcmo1E-F 5’-GGAATTCAGGAGCGGTTCCATGGAGATAATATTTG-3’
Bcmo1-R 5’-GACTAGTTAAAGACTTGAGCCACCATG-3’
SR-F 5’-GACTAGTGAATTCAGGAGGTAATAAATATGG-3’
SR-R 5’-CACTAGTTAGTTTTTGATTTTG-3’
a
Template binding regions are indicated with bold letters, start and stop codons are underlined, and restriction sites are double underlined.

arabinose, as the main and auxiliary carbon sources,
were added at concentrations of 0.5% to 2% (w/v) and
0.2% (w/v), respectively. The addition of the auxiliary
carbon source arabinose has been reported to increase
b-carotene production [15]. Glucose, galactose, xylose
and maltose were compared to glycerol as carbon
sources for retinoid production. Ampicillin (100 μg/mL)
and chloramphenicol (50 μg/mL) were added to the cul-
ture as required. Cell culture was carried out in a test
tube containing 7 mL of medium, and growth was
determined by measuring the optical density at 600 nm
(OD 600 ). For the two-phase culture for retinoid
production, 1 mL of dodecane (Cat. No. 297879, Sigma,
USA) was layered over 5 mL of culture medium.
Gene cloning and plasmid construction
The plasmids and PCR primers used in this study are
listed in Table 1. Common procedures, including geno-
mic DNA preparation, restriction digests, transforma-
tions, and other standard molecular biological
techniques, were carried out as described in the litera-
ture (Sambrook and Russell 2001). PCR was performed
using pfu DNA polymerase (Solgent Co., Korea) with a
standard protocol. The pBluescript, pTrc99A, and
Jang et al. Microbial Cell Factories 2011, 10:59
http://www.microbialcellfactories.com/content/10/1/59
pSTV28 plasmids were used for gene cloning and gene
expression (Table 1). The pS-NA plasmid containing the
operon for the MVA pathway was used as described
previously (Yoon et al., 2009). The BCM(D)O genes blh
and brp, brp2, and bcmo1 were amplified using PCR
from Halobacterium sp. NRC-1, Natronomonas pharao-
nis and Mus musculus, respectively. The PCR products
were cloned into the EcoRI and SpeI sites of pT-HB,
resulting in the retinal plasmids pT-HBblh, pT-HBbrp,
pT-HBbrp2 and pT-HBBcmo1 (Table 1). The blh gene
(Genbank accession number AAY68319) of the uncul-
tured marine bacterium 66A03 was synthesized by Gen-
ofocus (Daejeon, Korea) according to the codon-
optimization function of the company in-house software
for expression in E. coli. The synthetic gene named SR
(synthetic retinoid gene) was amplified using the PCR
primers SR-F and SR-R, and cloned into the EcoRI and
SpeI sites of pT-HB, resulting in pT-HBSR. The SR gene
cleaved from pT-HBSR with SpeI was cloned into the
corresponding site of pT-DHB, which resulted in pT-
DHBSR.
Analysis of b-carotene and retinoids
b-Carotene and retinoids were extracted from bacterial
cell pellets with acetone [13]. In the two-phase culture
system with a dodecane overlay, the upper dodecane
phase containing the retinoids was collected and centri-
fuged for 10 min at 14,000 rpm to remove all cellular
particles. The acetone extracts and dodecane phases
were analyzed with HPLC (LC-20A, Shimadzu, Kyoto,
Japan) at detection wavelengths of 370 nm (retinal), 340
nm (retinol and retinyl acetate), and 454 nm (b-caro-
tene) and using the Symmetry C18 (250 mm × 4.6 mm,
5 μm) with Sentry Guard C18 (15 mm × 4.6 mm, 5 μm)
HPLC columns (Waters, Milford, USA). The mobile
phases were 95:5 and 70:30 methanol and acetonitrile
for the retinoid and b-carotene analyses, respectively. A
flow rate of 1.5 ml/min and column temperature of 40°
C were applied for the HPLC analysis. Retinal (Cat. No.
R2500), retinol (Cat. No. R7632), retinyl acetate (Cat.
No. R4632) and b-carotene (Cat. No. C4582) were pur-
chased from Sigma (USA), dissolved in acetone, and
used as standard compounds. The results are presented
in the means ± SD from three independent experiments.
Additional material
Additional file 1: Retinoid production of various E. coli strains.
Retinoid production and cell growth of various E. coli strains harboring
pT-DHBSR and pS-NA. Culture was carried out in 2YT medium containing
0.5% (w/v) glycerol and 0.2% (w/v) arabinose for 48 hours at 29°C.
Retinal, retinol, and retinyl acetate are indicated with light gray, dark
gray, and black, respectively. For cell growth, the host strains are
indicated as symbols; MG1655, squares; DH5a, circles; XL1-Blue, triangles;
S17-1, reversed triangles; BL21, diamonds.
Page 11 of 12
Additional file 2: Effect of working volume on retinoid production.
Effect of working volume on retinoid production and cell growth of E.
coli harboring pT-DHBSR and pS-NA. The cultures were carried out in 2YT
medium containing 0.5% (w/v) glycerol and 0.2% (w/v) arabinose for 48
hours at 29°C. Retinal, retinol, and retinyl acetate are indicated with light
gray, dark gray, and black, respectively. For cell growth, the working
volumes are indicated as symbols; 3 mL, squares; 5 mL, circles; 7 mL,
triangles; 10 mL, reversed triangles.
Additional file 3: Effect of cultivation temperature on retinoid
production. Effect of cultivation temperature on retinoid production and
cell growth of E. coli harboring pT-DHBSR and pS-NA. The culture was
carried out in 2YT medium containing 0.5% (w/v) glycerol and 0.2% (w/v)
arabinose for 48 hours. Retinal, retinol, and retinyl acetate were indicated
with light gray, dark gray, and black, respectively. For cell growth,
temperatures are indicated as symbols; 29°C, squares; 34°C, circles; 37°C,
triangles.
Additional file 4: Effect of carbon sources on retinoid production.
Effect of carbon sources on retinoid production and cell growth of E. coli
harboring pT-DHBSR and pS-NA. Culture was carried out in 2YT medium
containing 0.2% (w/v) arabinose and 0.5% (w/v) glycerol, glucose, xylose,
maltose, or galactose for 48 hours at 29°C. Retinal, retinol, and retinyl
acetate are indicated with light gray, dark gray, and black, respectively.
For cell growth, the carbon sources are indicated as symbols; none,
squares; glycerol, circles; glucose, triangles; xylose, reversed triangles;
maltose, diamonds; galactose, stars.
Additional file 5: Effect of dodecane overlay on b-carotene
production. Effect of the dodecane overlay on b-carotene production
and cell growth of E. coli harboring pT-DHB and pS-NA. Culture was
carried out in 2YT medium containing 0.5% (w/v) glycerol and 0.2% (w/v)
arabinose with 1 mL of dodecane layered over 5 mL of culture broth for
48 hours at 29°C. Open bars and solid bars represent 24 and 48 hours,
respectively.
Acknowledgements
This work was supported by a grant (2009-0084490) from the Basic Research
Program, a grant (NRF-2010-C1AAA001-0029084) from the National Research
Foundation, MEST, and a grant from the Next-Generation BioGreen 21
Program (No. 2010-0000), Rural Development Administration, Korea. HJ Jang
and JH Kim are supported by scholarships from the BK21 Program, MEST,
Korea.
Author details
1
Division of Applied Life Science (BK21 Program), PMBBRC, Gyeongsang
National University, Jinju 660-701, Korea. 2Korea Research Institute of
Bioscience & Biotechnology, 52 Eoeun-dong, Yuseong-gu, Daejeon, Korea.
3
Department of Bioscience and Biotechnology, Konkuk University, Seoul 143-
503, Korea.
Authors’ contributions
SWK initiated and coordinated the project; HJJ and SHY performed the
research and wrote the paper; HKR, JHK, and CLW analyzed the data; JYK
and DKO reviewed the paper. All authors approved the final manuscript.
Competing interests
The authors declare that they have no competing interests.
Received: 26 May 2011 Accepted: 29 July 2011 Published: 29 July 2011
References
1. Sorg O, Antille C, Kaya G, Saurat JH: Retinoids in cosmeceuticals. Dermatol
Ther 2006, 19(5):289-296.
2. Amer M, Maged M: Cosmeceuticals versus pharmaceuticals. Clin Dermatol
2009, 27(5):428-430.
3. Mercier , Chabardes : Organometallic chemistry in industrial vitamin A
and vitamin E synthesis. Pure & Appl Chem 1994, 66(7):1509-1518.
4. Peck RF, Echavarri-Erasun C, Johnson EA, Ng WV, Kennedy SP, Hood L,
DasSarma S, Krebs MP: brp and blh are required for synthesis of the
Jang et al. Microbial Cell Factories 2011, 10:59 Page 12 of 12
http://www.microbialcellfactories.com/content/10/1/59

retinal cofactor of bacteriorhodopsin in Halobacterium salinarum. J Biol 26. Tsuruta H, Paddon CJ, Eng D, Lenihan JR, Horning T, Anthony LC,
Chem 2001, 276(8):5739-5744. Regentin R, Keasling JD, Renninger NS, Newman JD: High-level production
5. Sabehi G, Loy A, Jung KH, Partha R, Spudich JL, Isaacson T, Hirschberg J, of amorpha-4,11-diene, a precursor of the antimalarial agent artemisinin,
Wagner M, Beja O: New insights into metabolic properties of marine in Escherichia coli. PLoS One 2009, 4:(2):e4489.
bacteria encoding proteorhodopsins. PLoS Biol 2005, 3(8):e273. 27. Ramos JL, Duque E, Gallegos MT, Godoy P, Ramos-Gonzalez MI, Rojas A,
6. Martinez A, Bradley AS, Waldbauer JR, Summons RE, DeLong EF: Teran W, Segura A: Mechanisms of solvent tolerance in gram-negative
Proteorhodopsin photosystem gene expression enables bacteria. Annu Rev Microbiol 2002, 56:743-768.
photophosphorylation in a heterologous host. Proc Natl Acad Sci USA 28. Yoon KW, Doo EH, Kim SW, Park JB: In situ recovery of lycopene during
2007, 104(13):5590-5595. biosynthesis with recombinant Escherichia coli. J Biotechnol 2008,
7. Lee PC, Schmidt-Dannert C: Metabolic engineering towards 135(3):291-294.
biotechnological production of carotenoids in microorganisms. Appl 29. Baker ME: Evolution of mammalian 11beta- and 17beta-hydroxysteroid
Microbiol Biotechnol 2002, 60:(1-2):1-11. dehydrogenases-type 2 and retinol dehydrogenases from ancestors in
8. Rohmer M, Knani M, Simonin P, Sutter B, Sahm H: Isoprenoid biosynthesis Caenorhabditis elegans and evidence for horizontal transfer of a
in bacteria: a novel pathway for the early steps leading to isopentenyl eukaryote dehydrogenase to E. coli. J Steroid Biochem Mol Biol 1998, 66(5-
diphosphate. Biochem J 1993, 295(Pt 2):517-524. 6):355-363.
9. Lange BM, Rujan T, Martin W, Croteau R: Isoprenoid biosynthesis: the 30. Kuznetsova GM, Kartasheva ZS, Kasaikina OT: Kinetics of limonene
evolution of two ancient and distinct pathways across genomes. Proc autooxidation. Russian Chemical Bulletin 1996, 45(7):1682-1685.
Natl Acad Sci USA 2000, 97(24):13172-13177. 31. Knight AET, (NJ), Talati, Akshay (East Meadow, NY), Zecchino, Jules (Closter,
10. Martin VJ, Pitera DJ, Withers ST, Newman JD, Keasling JD: Engineering a NJ): Stable compositions containing biologically active components.
mevalonate pathway in Escherichia coli for production of terpenoids. United States: E-L Management Corp. (New York, NY); 1999.
Nat Biotechnol 2003, 21(7):796-802. 32. Shen CR, Lan EI, Dekishima Y, Baez A, Cho KM, Liao JC: High titer
11. Vadali RV, Fu Y, Bennett GN, San KY: Enhanced lycopene productivity by anaerobic 1-butanol synthesis in Escherichia coli enabled by driving
manipulation of carbon flow to isopentenyl diphosphate in Escherichia force. Appl Environ Microbiol 2011.
coli. Biotechnol Prog 2005, 21(5):1558-1561. 33. Bond-Watts BB, Bellerose RJ, Chang MCY: Enzyme mechanism as a kinetic
12. Yoon SH, Lee YM, Kim JE, Lee SH, Lee JH, Kim JY, Jung KH, Shin YC, control element for designing synthetic biofuel pathways. nature
Keasling JD, Kim SW: Enhanced lycopene production in Escherichia coli chemical biology 2011.
engineered to synthesize isopentenyl diphosphate and dimethylallyl
diphosphate from mevalonate. Biotechnol Bioeng 2006, 94(6):1025-1032. doi:10.1186/1475-2859-10-59
13. Yoon SH, Lee SH, Das A, Ryu HK, Jang HJ, Kim JY, Oh DK, Keasling JD, Cite this article as: Jang et al.: Retinoid production using metabolically
Kim SW: Combinatorial expression of bacterial whole mevalonate engineered Escherichia coli with a two-phase culture system. Microbial
pathway for the production of beta-carotene in E. coli. J Biotechnol 2009, Cell Factories 2011 10:59.
140:(3-4):218-226.
14. Wang C, Yoon SH, Shah AA, Chung YR, Kim JY, Choi ES, Keasling JD,
Kim SW: Farnesol production from Escherichia coli by harnessing the
exogenous mevalonate pathway. Biotechnol Bioeng 2010, 107(3):421-429.
15. Yoon SH, Park HM, Kim JE, Lee SH, Choi MS, Kim JY, Oh DK, Keasling JD,
Kim SW: Increased beta-carotene production in recombinant Escherichia
coli harboring an engineered isoprenoid precursor pathway with
mevalonate addition. Biotechnol Prog 2007, 23(3):599-605.
16. Leuenberger MG, Engeloch-Jarret C, Woggon WD: The Reaction
Mechanism of the Enzyme-Catalyzed Central Cleavage of beta-Carotene
to Retinal This research was supported by F. Hoffmann-La Roche AG
and the Swiss National Science Foundation. We are grateful to F.
Hoffmann-La Roche AG for a generous gift of carotenoids and Dr. Claus
Bornemann for preliminary experiments. Angew Chem Int Ed Engl 2001,
40(14):2613-2617.
17. Borowski T, Blomberg MR, Siegbahn PE: Reaction mechanism of
apocarotenoid oxygenase (ACO): a DFT study. Chemistry 2008,
14(7):2264-2276.
18. Liden M, Eriksson U: Understanding retinol metabolism: structure and
function of retinol dehydrogenases. J Biol Chem 2006,
281(19):13001-13004.
19. Lin M, Zhang M, Abraham M, Smith SM, Napoli JL: Mouse retinal
dehydrogenase 4 (RALDH4), molecular cloning, cellular expression, and
activity in 9-cis-retinoic acid biosynthesis in intact cells. J Biol Chem 2003,
278(11):9856-9861.
20. Ross AC: Retinol esterification by rat liver microsomes. Evidence for a
fatty acyl coenzyme A: retinol acyltransferase. J Biol Chem 1982,
257(5):2453-2459. Submit your next manuscript to BioMed Central
21. Bempong DK, Honigberg IL, Meltzer NM: Normal phase LC-MS and take full advantage of:
determination of retinoic acid degradation products. J Pharm Biomed
Anal 1995, 13(3):285-291.
• Convenient online submission
22. McCaffery P, Evans J, Koul O, Volpert A, Reid K, Ullman MD: Retinoid
quantification by HPLC/MS(n). J Lipid Res 2002, 43(7):1143-1149. • Thorough peer review
23. Swindell EC, Eichele G: Retinoid metabolizing enzymes in development. • No space constraints or color figure charges
Biofactors 1999, 10:(2-3):85-89.
24. Perlmann T: Retinoid metabolism: a balancing act. Nat Genet 2002, • Immediate publication on acceptance
31(1):7-8. • Inclusion in PubMed, CAS, Scopus and Google Scholar
25. Kim SW, Keasling JD: Metabolic engineering of the nonmevalonate
• Research which is freely available for redistribution
isopentenyl diphosphate synthesis pathway in Escherichia coli enhances
lycopene production. Biotechnol Bioeng 2001, 72(4):408-415.
Submit your manuscript at
www.biomedcentral.com/submit